Haldane and the Emergence of Modern Evolutionary Theory 61survival is approximately equal to 2k, and thus remains finite even in a large
population. If the mutation is recessive, and the selective advantage that it confers
is stillk, the probability of its survival is
√
K
N, whereNis the size of the population.
Thus, the chance of survival decreases with the size of the population. Once
again, inbreeding or self-fertilization brings this probability back to the level of
the dominants.
If mutation occurred with an appreciable frequency, the situation was entirely
different. A series of simple calculations showed that “any advantageous or not
too disadvantageous factor will certainly be established” [1927b, 840] in the pop-
ulation. Haldane was impressed with the power of mutation:
if selection acts against mutation, it is ineffective provided that the
rate of mutation is greater than the coefficient of selection. Moreover,
mutation is quite effective where selection is not, namely in causing
an increase of recessives where these are rare. It is also more effective
than selection in weeding out rare recessives provided that it is not
balanced by back mutation of dominants. Mutation therefore deter-
mines the course of evolution as regards factors of negligible advantage
or disadvantage to the species. [1927b, 842]The only caveat was that it “can only lead to results of importance when its
frequency becomes large” (p. 842). Haldane would never renege on this rather
optimistic assessment of the evolutionary significance of mutations.^21 Six years
later, when he returned to a systematic — but highly speculative — treatment
of the topic, he invoked mutation to explain the disappearance of useless organs,
recapitulation and the observation that the heterogametic sex is usually the male
[Haldane, 1933].^22 However, Haldane’s most important application of the results
of this paper was his later use of the balance between selection and mutation to
provide the first estimate of a human mutation rate [Haldane, 1935].
Three years later, at the end of 1929, he turned to a systematic analysis of
isolated populations with immigration. Ten different sets of models were treated in
Part VI [Haldane, 1930]. These models assumed that, in the isolated population, a
new type is favored (with selection coefficientk), while migrants of the original type
continue to enter the population. In each generation, the number of immigrants
wasltimes the total isolated population. In each model unlessk/lexceeded a
certain critical value, as should perhaps be expected, immigration swamped out
the original type. However, even ifk/ldid exceed that value, for the selected
type to survive, it was sometimes necessary that its original frequency was high.
When both these conditions were met, the population would reach an equilibrium
composition with both types present. However, some of these equilibria were
unstable: a fluctuation would drive one of the types to fixation.
(^21) This is seen, for instance, inThe Causes of Evolution[Haldane, 1932].
(^22) Meanwhile, a consideration of the secondary effects of mutations led him into the burgeoning
controversy, mainly between Fisher and Wright, over the evolution of dominance — Falk [2001].